This is a LoRA that is an extraction from a Base Model that I created called Princesses. The "sd_xl_base_1.0_0.9vae.safetensors" Base Model was used as the subtraction from the Princesses Base Model, so the LoRA generates the same results as the Princesses Base Model when used with the "sd_xl_base_1.0_0.9vae.safetensors" Base Model. It works best using only a specific positive prompt of what you want to generate and the negative prompt that is shown in the workflow, with the LoRA strength, set at 1.0, and a {randomize} sampler setting.